The Design Of Home Private Cloud System Based On Docker

Nowdays, with the rising of mobile devices, a large amount of data generated by such devices everyday. Cloud computing meets the user's need of data processing, the small or medium enterprises and individuals use Cloud computing to provide data storage and computing services. Cloud computing saves the storage and calculation cost. But everything has two sides,cloud computing brings convenience,and at the same time also has certain risk.Most of people store their data in the public cloud, and in the public cloud system, people share the same computing resources, people can't exactly know where the controllable resources run, it likely to have the risks of being used illegally by other people. As the amount of personal data is bigger and bigger,people are paying more attention to the cloud data security issues. Though the traditional encryption method can alleviate a certain amount of crisis of cloud computing, the current Social Engineering database is more and more rampant.The risk of leaking people's accounts has been higher and higher. The problem of personal cloud data security is complexed, building a private cloud system, it is the most basic measure to solve this problem from the origin.When it comes to cloud computing technology, the deployment is very difficult. Generallly speaking,only a few big companies can provide such service. Amazon provides a VPC services, which is a virtual private cloud service, but its essence is a kind of public clouds,still cannot reduce the security of the data. In 2014, an operating system-level virtualization technology is very popular.Its name is Docker, and greatly reduce the hardware cost, make cloud computing deployment can be done by the individual completely, the deployment and reconstruct of a system become simpler and more efficient.This paper is based on the analysis of cloud security risk, and then puts forward a design scheme for family private cloud, which is based on Docker. Firstly, explaining the concept of virtualization and cloud computing technology in detail, and then makes a thorough study of the operating system virtualization technology, which used Linux kernel technology, such as Namespace, Cgroup, AUFS and Devicemapper. Finally, making a detailed analysis of the Docker's structure, understanding the Docker container in essense.Docker technique is used to implement the virtualization in the cloud computing. It is the first step to building a private cloud.Then using the distributed storage system, Moose FS, to perform high performance cloud storage.When I check Moose FS source code, I find the load balancing algorithm have the problem of low node utilization rate. Then I put forward a new improved algorithm, this algorithm can make the data storage more centrally, convenient for the user performs data backup and migration, and improve the utilization rate of storage nodes, then make a test to verify this algorithm.On the basis of analyzing the Docker and optimizing the Moose FS, this paper combines with the redevelopment of Seafile storage management system,completes the construction of the whole family private cloud system, and then transplants related image files.Then constructing the images of Moose FS distributed storage system for its various nodes, building Seafile service image. Finally, the private cloud host runs six Docker containers, every container runs its corresponding image.and then make a comprehensive test to the system.